航拍数字目标图像跟踪精确匹配仿真

摘要

This article proposes an accurate matching method for tracking of digital target image of aerial photo based on direction and gradient of edge.Firstly,the author used active shape model to build information model for direction and strength of image pixel edge and extracted shape outline of the digital target image,then emloyed inference theory to select connected region describer of the digital target image and calculated fuzzy matching degree among each connected region of image.Moreover,the research fed back and amended matching relation of feature point among the connected region.Experimental results show that the method has high matching precision.It can improve identification quality of the digital target image of aerial photo.%对航拍数字目标图像的跟踪精确匹配,能够提升对航拍数字目标图像的识别质量.对数字图像进行跟踪匹配,需要计算图像各连通区域间的模糊匹配度,反馈修正图像连通区域间特征点的匹配关系,完成图像跟踪的精确匹配.传统方法先对图像的角点进行检测,引入图像旋转特征,但忽略了对得到的特征进行匹配,导致图像跟踪匹配的准确性较低.提出基于边缘方向和梯度的航拍数字目标图像跟踪精确匹配方法.利用主动形状模型理论建立图像像素边缘方向和强度的信息模型,提取航拍数字目标图像形状轮廓,推理思想选择航拍数字目标图像连通区域描绘子,计算图像各连通区域间的模糊匹配度,反馈修正图像连通区域间特征点的匹配关系,以此为依据完成对航拍数字目标图像跟踪精确匹配.实验结果表明,所提方法匹配精度高,大幅度提升了航拍数字目标图像的识别质量.